RoboJackets’ RoboRacing outscored all other teams at the International Autonomous Robot Racing Challenge (IARRC) July 11-14 in Waterloo, Canada, to bring home a first place win and $3000 CAD.
"This was an incredible win for us because we completely redesigned our software stack since the 2018 competition to a more traditional computer vision approach," said Brian Cochran, senior software team member.
"A lot of things we do in RoboJackets we learn here, before we learn it in the classroom, " explained Cochran, a computer science 2020 major.
IARRC is focused on high speed autonomous racing compared to solely self-driving competitions.
Competition consisted of four events- a straight-line drag race, a winding circuit race, an obstacle course, and an urban challenge where the vehicle must identify road signs and keep within its lane on a two-lane road.
The vehicle, named Sedanii, is built from an off the shelf 1/6 scale RC car with custom framed components added to mount three cameras, nearly 140 watt hours of batteries, a quad core mini desktop computer, and an inertial measurement unit (IMU), and a motor encoder.
Sedanii fuses data from the three cameras to create a map of its surroundings containing locations of course boundaries and obstacles.
The car plans a path through the map of obstacles utilizing a vehicle kinematic model to predict and optimize its path while maximizing speed and remaining in control. This path then gets translated into commands which are sent to the vehicles speed and steering controllers.
This was RoboJackets’ fifth appearance at IARRC and their third win. In the past they have also competed in the Sparkfun Autonomous Vehicle Challenge, and intend on branching out into autonomous go-kart racing this coming year.
RoboJackets and the other Georgia Tech student competition teams are open to students of all majors, backgrounds, and levels of experience. New members are welcome.
